The Echo of Knowledge Production in Historical Sound Documents by Anette Hoffman

21 May 2020 - 12:00
| Kxara the younger with phonograph (R. Pöch, 1908)


The starting point for my book Kolonialgeschichte Hören: Das Echo gewaltsamer Wissensproduktion in historischen Tondokumenten aus dem südlichen Afrika (Hearing Colonial History: The Echo of Violent Knowledge Production in Historical Sound Documents from Southern Africa)is my engagement with colonial history as a listener. By means of listening to the persistent echo of colonial knowledge production in (mostly) linguistic recordings from southern Africa, I came to hear these recordings not as the language examples as which they were recorded and archived, but as acoustic sources of colonial history. This understanding of collections of historical recordings extends the notion of the colonial archive, which then includes spoken and sung texts and performative gestures that only appear in acoustic documents.

Once one turns to listening, historical audio collections resound with technological achievements entwined with the idea that languages and cultural expressions need to be salvaged. Attention to connected documentation shows that researchers who (said they) worried about the disappearance of languages, types of music and oral repertoires often exploited opportunities that followed from the violent subjugation of colonised populations.

The first part of the book approaches the specificities of sound archives as historical sources. Engaging with these acoustic documents, I listen to crackling shellac records and wax cylinders, on which voices can be heard, which once spoke, commented, whispered, narrated, sung, and criticised. The recordings I have studied in the last decade were archived by linguists, anthropologists, adventurers, and missionaries. Some of them were recorded a century ago, all of them were created in asymmetric, yet collaborative, situations of knowledge production. Today, most of these recordings are sequestered in archives where they have been received and configured as specimens; they are filed as examples of languages and music (Lange 2015; 2018; Hoffmann 2015; 2018; 2019). So far, textual contents of recorded speech and song rarely surface in the catalogues of sound archives, the contents are not searchable in archival registers, and the recordings rarely enter the public sphere. Speakers, who were seen as ‘native informants’, were systematically absented from the archival catalogues. The collaborative nature of knowledge production is perpetually concealed, and sound archives have kept quiet about the colonial, violent context of their making. This figuring of acoustic collections is durable; it has shaped the recordings and indelibly watermarked their form and content (Hamilton et al. 2002; Stoler 2009; 2016; Quijano 2007). Still, historical sound recordings allow interested listeners to revisit acoustic traces that resonate with the very moment of their making. While the recorded voice is mediated, what may be heard in a recording often differs from what can be read in transcriptions, or from what appears in the catalogue of the archive.

In this book I discuss historical recordings as potentially speaking in ways that differ from what can be seen in the photographs of the same collections, or from what one reads in the scholarly texts of the so-called collectors, or the travelogues of anthropologists. The book argues for expanding our understanding of the colonial archive toward the inclusion of acoustic documents, which carry the echo of colonial knowledge production. Subsuming sound collections within the colonial archive at large is a strategic move in at least two senses: it argues for the inclusion of historically pertinent, as yet barely known acoustic documents, adding recorded spoken or sung texts to the debate around who speaks and whose/which histories can be found in the colonial archive.  Attending to historical voice recordings and music as semantically and performatively meaningful utterances regularly allows for the surfacing of speaking positions, in which the comments or narratives of those who were subjected to the study, for instance of anthropologists, become audible (Hoffmann 2014). This often alters the themes to which collections speak, sometimes directing my attention towards topics I could not have preconceived. 

The chasm between what is catalogued and what is audible appears in the collection of the Austrian anthropologist Rudolf Pöch, which was recorded in the Kalahari in 1908 as part of an attempt to document ‘Bushman languages’. Ins Archiv hören engages with this specific collection of recordings with (mostly) Naro speakers, which has been published as a CD, titled Rudolf Pöch’s Kalahari Recordings (1908).  As ever so often, the title of the CD identifies the anthropologist as the sole creator of the recordings.The collection of recordings entails acoustic documents in a variety of archival conditions: some are documented in writing, whereas others are not, Pöch’s transcriptions of the recordings are mostly doubtful. Due to the poor sound quality of the recordings many of the translations, which Pöch provided, cannot not be re-translated. Some of the speakers are traceable across different parts of this collection, appearing on photographs, or film, or featuring in Pöch’s diaries. Quite important for the question in which ways sound recordings can speak to practices of colonial knowledge production and the plundering that has been called collecting, is the fact that these acoustic recordings are part of a vast, disturbing accumulation of objects. These objects were amassed by the Austrian anthropologist, on his journey through the Kalahari, during the colonial war in Namibia (then German Southwest Africa) in 1908-9, and in South Africa and Botswana (then British Bechuanaland) around the same time. The collection consists of more than thousand items that then became ethnographic objects, but also 1500 photographs, film, 68 acoustic recordings, Pöch’s field notes, parts of animals and plants, and the human remains of originally 171 people, many of whom were farmworkers (Plankensteiner 2009; Rassool & Legassick 2000; Schasiepen 2019; Pöch 1908; 1909, 1910). Engaging with some of the recordings of this collection, I have tried to make sense of what was said by four Naro speakers and the anthropologist himself, who in the documentation of the CD is falsely registered as Bushman.

Three main practices form the basis of my interpretation of the recording: translation, close listening and re-assembling.

I was immensely lucky that Job Morris, activist for the rights of San from D’kar in Botswana (the area were the recordings were produced more than 110 years ago) agreed to translate the recordings. Without his engagement with these recordings no new interpretation of the Naro recordings would have been possible. Translation in this case meant listening to the recordings many times to accomplish an interpretation of tone, tempo and manner of speech to understand the meaning of utterances.

What I call ‘close listening’ 4 describes the attempt ‘to know by ear’, that is, to grasp as much as possible of the audible features of a recording. This includes attention to sonic features which do not appear on the label, for instance, the sound of the pitch pipe (which indicates the speed at which the recording should rotate), the noise of a rotating cylinder or scratched record, (which can deliver clues on how often the record has been played) the recordist’s announcement (the ‘acoustic tag’), the languages, performative and musical genres documented on the recording, the features of the voice of the speaker(s) and singer(s), together with accent, pauses, sounds that give away the presence of an audience, and background noises. Close listening is a collaborative practice, it means listening with the translators, with specialists on the language, the history, the cultural background of those who were recorded. It also means to acknowledge that what has been created in a collaborative moment in the past cannot be interpreted in the present by a (Western) researcher or archivist alone.

Apart from exposing that what can be heard on a recording may or may not be found on the file’s label or its written documentation, this exercise in listening makes audible that archived sound files often speak beyond the object status attributed to them by the recordists and in the archival documentation. Listening to all that is audible on the sound file makes clear that most recordings are composite acoustic objects that can be heard in different ways.

Close listening also alerts to the positionality of listening in time and space, as well as to the status of sound in a given environment or situation: What one hears or does not hear, what a listener attends to or tends to ignore, is intrinsically related to his/her own background, to the languages spoken, the practices of speaking and knowledge transmission one knows, and the expectations listeners have. Modes of hearing and listening are culturally and historically informed. They precipitate in languages and philosophies, have histories and impacts in the present (Baumann 1997; Chernoff 1997).

Sound archives keep what has been recorded for specific purposes and interests; attentive listening may perceive that the rationale of the researchers did not entirely control what entered the archive. The excess of voice vis-à-vis the spoken or sung words – in pronunciation, tone, and the ability to express non-verbally –  at times evaded the registration of the archiving process. Thus, the performativity, meaning and messages of recorded sounds, words, songs, stories, or even example sentences are not always contained within the rules and practices of recording.

Marked by the archival politics that have produced them in the first instance, historical recordings often present a neuralgic nexus in archiving, because they constitute an interface of different cultural practices of conservation or archiving that become articulated – in the double sense of expressing and joining – by means of recording. What has been archived as examples of music and languages often holds elements of repertoires, which can be fragments of record keeping such as oral poetry, or songs and narratives that are part of a body of historiology.5  This means that the expressiveness of recorded voices and sounds benefit immensely from knowing the situation and the politics of their production as recordings. Additionally, the complexity of oral genres, or the researchers’ inability to understand languages means that contents of various repertoires may have entered sound archives unidentified. What was said was often not understood by those who recorded. Often the contents of speech were outside of the field of interest in which the archivists operated, or the double meaning or performative sense of songs escaped the radar of regulation and censorship.

The third major element of my interpretation of historical sound recordings is re-assembling (Harris 2013). Regarding this specific sound collection this entails a systematic reconnection of the dispersed objects of collections, and, if possible, the reconstruction of situations of recording. In this way photographs, objects, but also imperial power relations, the situation of Naro speakers during a severe drought and in a war, as well as Pöch’s practices of plundering with military support, speak to each other. The objects connected to Pöch’s travels in southern Africa are distributed to five archives and institutions in Vienna according to the logic of disciplines, and the politics of archiving. In this way the traces of the speakers have become fragmented:|Kxara the younger (he is one of two speakers with this name) appears on several recordings that are held by the Phonogrammarchiv in Vienna, and appear in the CD of 2003, his photographs are found in three different collections in Vienna, a cinematographic film that shows him is held at the Filmarchiv Austria. Identifying and re-assembling the objects that carry |Kxara’s traces rearrange these objects on other terms: not around the collector, but around a speaker. This reorganisation focuses on |Kxara’s role in the collaborative practice of knowledge production: he translated, cooked, looked after the trek oxen, steered the wagon, operated the phonograph while Pöch filmed the scene of recording (without |Kxara in the picture), and also spoke into the recording device. Close listening in combination with re-assembling pertinent documents and archival objects can significantly reframe acoustic documents and speaking positions, as well as add to the understanding of textual content.

The second chapter of the book engages with the disappearance of words that were recorded, documented in writing, but which nevertheless were lost irreversibly. This loss of (once recorded) words is not accidental. Instead, it is the result of a series of decision that systematically neglected the content of speech in favour of the spectacle of (then) new technologies and of recordings as specimens. I discuss this effect along Pöch’s cinematographic and phonographic recording of a speaker in Botswana, who was filmed while he spoke. The silent movie was synchronised with the original sound recording by Diedrich Schüller, the former director of the Phonogrammarchiv in Vienna, around 80 years later. The film clip circulates in the web titled ‘Bushman speaks into phonograph’ with no subtitles. It celebrates Pöch as a pioneer of ethnographic film, while silencing the speaker, who gesticulates without words, transmitting the (moving) images of a Bushman who cannot be understood. I discuss the recurring scene in which people were prompted to speak into the phonograph but not listened to as speakers with Rancières notion of mésentente,6  which helps to theorise the situation of speaking in the moment of colonial (for instance linguistic) knowledge production. Words were recorded and understood as language but were not heard as meaningful utterances, let alone as criticism or comments on the practices of knowledge production. They thus did not prompt a response or change in conduct. Mésentente, often translated as disagreement, here speaks of a conflict in the assertion of a situation of speaking, in which the speakers intended to communicate, often using the recording situation as an occasion to speak, whereas the researcher was not interested in the content because he (rarely she) had predefined the situation as an occasion to extract knowledge from a speakers. The inability, or unwillingness, to hear meaningful utterances was therefore not merely related to language barriers, but to the effect of an asymmetric relationship between researcher and speakers. This regular moment of mésentente has durable effects. It continues to operate until today, by omitting the words and comments from the registers of archives, by perpetuating to treat spoken content as irrelevant, and by not mentioning the speakers as authors of recordings. This means that the sharpest criticism, the most urgent pleas were filed as language examples and thus disappeared in the archives (Sonderegger 2009).

The third part of the book listens to the anthropologist who, in the written register of the CD, had become a Bushman. Pöch recorded his own voice in a place he called Kxau (Kg’au tshàa), in the vicinity of Ghanzi, British Betchuanaland (now Botswana, CD booklet: 36). On this recording, almost drowned out by crackle and hiss, one can hear his faint voice speaking with a thick Austrian accent in English, German and Afrikaans. The recording busts the myth of Pöch’s ability to communicate with the Naro speakers. On the recording the anthropologist speaks a bewildering mix of languages that even in the borderzone must have challenged the people he spoke to (Landau 2010). His question ‘Is julle bang (are you afraid)?’ becomes unheimlich (uncanny) in connection with the German Schutztruppe’s support of his expedition and his systematic desecration of graves to send human remains to Austria. No less disturbing is the infantilising way in which Pöch speaks to his assistant. On the recording the relationship of the anthropologist to his assistant becomes audible. The topic of fear appears again in the recording of |Kxara the younger, who angrily addresses the reluctance of Naro speakers to communicate with the anthropologist. The recurrence of the topic of fear, or caution, speaks of the omnipresent colonial violence and the precarious situation of Naro speakers in the area, which Pöch does not address.

In the last part of the book a recording with |Kxara the older (the second speaker with this name) leads into the storeroom of the Weltmuseum in Vienna: on one of his recordings, |Kxara demands that the anthropologist immediately returns his knife. He repeats his request no less than eight times. He also expresses his irritation about the behaviour of the anthropologist who refuses to share his resources with the Naro speakers whom he visits. |Kxara’s recordings thus deliver aspects of a critical evaluation of research ethics in Kg’au tshàa in 1908. |Kxara, who did not work for Pöch, obviously did not see himself in a subordinate position. Instead, he asked his knife back and criticised the anthropologist for his refusal to adapt to the sharing economy practiced among the Naro speakers. No response by the anthropologist appears in the documentation, Pöch’s transcription does not communicate this conflict. In the collection of the Weltmuseum, the knife could not be found. Yet following the speaker’s anger, a closer inspection of Pöch’s practices of plunder shows, that ‘collecting’ objects from Naro speakers was most successful, while the German military was present. In fact, Pöch amassed most of his collection of ‘Buschman Ethnografika’ while he stayed at the military station in Oas and during an expedition with camels and the Schutztruppe, in comparison to very few items that he was able get hold of during his stay with Naro speakers in Kg’au tshàa (Botswana).

Detail of photograph of the Kamelexpedition with the Schutztruppe (Pöch, 1908)

|Kxara’s recording shows the potential of sound archives as sources of alternative information about the practice of appropriating objects, which is relevant particularly in the ongoing debate on restitution of ethnographic collections. The former owner’s view might differ considerably from the documentation that can be found in the registers of European museums and collections.




Anette Hoffmann presented “The long echo of imperial knowledge production: Listening to phonographic recordings of Naro speakers from the Kalahari (1908),” based on this upcoming publication at the APC’s first research lab of the year, on Thursday 13 February, 2020. 

Bushman is an ethnographic stage name in Pöch’s writing on his journey. Under this name speakers appear whom he had pre-defined as the subjects of  his research. I have written on the term Bushman and its uses in my book, but cannot include this here.

3 Dietrich Schüller (Hg.) Rudolf Pöch’s Kalahari Recordings (1908). Sound Documents from the Phonogramarchiv of the Austrian Academy of Sciences. The Complete Historical Collections 1899–1950. Series 7, Wien: OAW 2003.

4 The notion of ‘close listing’ as a method of interpretation historical recordings first came up in conversations with Britta Lange, when we taught a seminar on sound archives at the Humboldt University together in 2012. It was developed further in workshops on ‘knowing by ear’, which I organised at APC in 2013 and 2014.

With historiology I mean orally construed and transmitted interpretations of history.

Jacques Rancière, Das Unvernehmen, Frankfurt a.M.: Suhrkamp 1995, S. 10f.; Ruth Sonderegger, „What One Does (Not) Hear. Approaching Canned Voices Through Rancière“, in: Anette Hoffmann (Hg.): What We See. Reconsidering an Anthropometrical Collection: Images, Voices and Versioning, Basel: Basler Afrika Bibliographien 2009, S. 58–84.



Baumann, Max Peter. 1997. Preface: Hearing and Listening in Cultural Context. The World of Music. 39 (2), Special Issue: Cultural Concepts of Hearing and Listening, 3-8.
Chernoff, John. 1997. Hearing in West African Idioms. The World of Music. 39 (2), Special Cultural Concepts of Hearing and Listening, 3-8
Hamilton, Carolyn, Verne Harris, Jane Taylor, Michele Pickover, Graeme Reid and Razia Saleh. 2002. Refiguring the Archive. Dordrecht/Boston/London.
Harris, Rodney. 2013. Reassembling Ethnographic Museum Collections. Rodney Harris/Sarah Byrne/ Anne Clarke (eds.) Reassembling the Collection: Ethnographic Museums and Indigenous Agency. Santa Fe 2013.
Hoffmann, Anette (ed.). 2009. What We See. Reconsidering an Anthropometrical Collection: Images, Voices and Versioning. Basel 2009.
Hoffmann, Anette. 2014. Verbal Riposte: Wilfred Tjiueza’s Performances of Omitandu as Responses to the Racial Model of Hans Lichtenecker. Hans D. Christ/Iris Dressler/Christine Peters (eds.) Acts of Voicing. Leipzig, 323-332.
Hoffmann, Anette. 2015. Listening to Sound Archives: Introduction to edited section. Social Dynamics 41(1) Special Section: Listening to Sound Archives, edited by Anette Hoffmann, 73–83.
Hoffmann, Anette 2018 „Kolonialität“ Daniel Morat & Hansjakob Ziemer (eds.) Handbuch Sound. Geschichte – Begriffe – Ansätze. Stuttgart: Metzler, 387-390.
Hoffmann, Anette 2019. ‘Achtung Aufnahme!’ Akustische Spuren der kolonialen Wissensproduktion” Iris Edenheiser and Larissa Förster (eds.) Museumsethnologie – Eine Einführung. Theorien – Debatten – Praktiken. Berlin: Dietrich Reimer Verlag, (in print).
Lange, Britta. 2015. Poste Restante, and Messages in Bottles: Sound Recordings of Indian Prisoners in the First World War in: Social Dynamics 41 (1), 2015: Special Section Listening to Sound Archives, S. 84-100.
Lange, Britta 2018. „Archiv“ Daniel Morat & Hansjakob Ziemer (eds.) Handbuch Sound. Geschichte – Begriffe – Ansätze. Stuttgart: Metzler, 236-240.
Plankensteiner, Barbara. 2009. ‘Auch hier gilt die Regel, Buschmanngut und Fremdgut auseinanderzuhalten’. Rudolf Pöchs Südafrika-Sammlung und ihre wissenschaftliche Bearbeitung durch Walter Hirschberg”, in: Archiv für Völkerkunde 59–60, 2009, S. 95–107.
Pöch, Rudolf. 1908. Berichte an die Akademie in: Anzeiger der Akademie der Wissenschaften, XLV. Jahrgang
Pöch, Rudolf 1909. Berichte an die Akademie in: Anzeiger der Akademie der Wissenschaften, XLVI. Jahrgang.
Pöch, Rudolf. 1910. Reisen im Innern Südafrikas zum Studium der Buschmänner in den Jahren 1907-1909. Zeitschrift für Ethnologie, 42 (2), 357-361.
Quijano, Aníbal. 2007. Coloniality and Modernity/Rationality. In: Cultural Studies 21(2–3), 168–178.
Rancière, Jacques. 1995. Das Unvernehmen, Frankfurt a.M.
Rassool, Ciraj and Martin Legassick. 2000. Skeletons in the Cupboard: South African Museums and Human Remains, 1907-1917. Cape Town/Kimberley.
Schasiepen, Sophie. 2019. Die ‚Lehrsammlungen’ von Dr Rudolf Pöch an der Universität Wien. Anthropologie, Forensik und Proveninienz in: Zeitschrift für Kulturwissenschaften 1, S. 15–28.
Schüller, Dietrich. 2003. Rudolf Pöch’s Kalahari Recordings (1908). Wien.
Sonderegger, Ruth. 2009. „What One Does (Not) Hear. Approaching Canned Voices Through Rancière“, in: Anette Hoffmann (Hg.): What We See. Reconsidering an Anthropometrical Collection: Images, Voices and Versioning, Basel
Stoler, Ann Laura. 2009. Along the Archival Grain: Epistemic Anxieties and Colonial Common Sense. Princeton.
Stoler, Ann Laura. 2016. Duress. Imperial Durabilities in Our Time. Durham.